Epidemiology Study Of Occupational Stress Assessment And Its Health Effect

With social development and economic globalization, more and more employees are experiencing job stress since organization reformed and new policy took into effect at recent years. Long-term excessive work pressure contributed to occupational stress, which is becoming a sever problem of public health. In 1994, the United Nations report had pointed out "World is becoming into brimming stress one", which based on International Labor Organization investigation. As Theorell commented "Work stress is the making of modern epidemic." Occupational stress has regarded as a new occupational hazard factor and affected and impaired people's health widely. It is a foundation issue how to assess occupational stress. Moreover, it is an urgent task for us to develop a good instrument of occupational stress in China. At the same time, we are short of convincing evidence data to show what and how level of occupational stress affect human health. Purpose of the study is to develop a tool for occupational stress assessment, and to explore its psychological and physical effect. It would be of great importance to indicate job stress scale, and provide academic basis the prevention of occupational stress injury.A cross-sectional study was employed for this part of the study. There were seven occupational groups selected as subjects based on job stress types such as primary and middle school teacher, administer officer and professional, foreign manufacturing corporation employees, serving company employees, traffic plice, community health center staff and estate management employees. All participants consented and completed questionnaire by self-administer. The health effects of occupational stress were assessed by self-report and physical examination. This study started with occupational stess questionnaire assessment and identified health effects of occupational stress.Part I Study into Occupational Stress Assessment1 Reliability and validity of occupational stress questionnaireThe concise occupational stress questionnaire was designed according to JobDemand-Control (JDC) Model developed by Kasarek and Effort-Reward Imbalance (ERI) model developed by Siegrist after a great deal of literatures has been reviewed. There were 1368 questionnaires qualified. All 24-item from JDC model questionnaire was subjected to explore factor analysis, and five factors were identified, which explained 61.1% of total variance. Its Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) reached to 0.872. All 23-item from ERI model questionnaire was subjected to explore factor analysis, and four factors were identified, which explained 58.3% of total variance. Its KMO reached to 0.927. Finding showed Cronbach's alpha were 0.69-0.90 among latent factors and 0.88 total items of occupational stress questionnaire through the consistent test. All 44 items with eight latent factors from whole questionnaire identified by confirm factor analysis from structure equation model, which was a better one by absolute model fit index and Root Mean Square Error of Approximation (RMSEA) was only 0.063.2 Comparison occupational stress scale between different occupationsDiscriminant validity is an important basis in weighting the quality of concise occupational stress questionnaire. Discriminant capability between occupations served as an important index of questionnaire quality because there is lack of gold standard in occupational stress. Finding showed there was significant different between every factors' score of questionnaire at seven occupations by ANOVA. There were 71.9% objectives with high occupational stress by JDC model and 17.9% persons with high stress by ERI model. There was a significant difference between different occupations by JDC model or by ERI model. There was a good consistent for stress level among occupations by JDC model and ERI model. It is a better instrument of occupational stress assessment with better reliability and validity.3 Reliability and validity of Job Burnout QuestionnaireA job burnout questionnaire in Chinese was developed according to Maslach Burnout Inventory with 19-items revised by Boles. The 19-items questionnaire was subjected to explore factor analysis, and three factors were identified, which explained 58.7% of total variance. Its KMO reached to 0. 877. Finding showed Cronbach's alpha were 0.84～0.87 among latent factors and 0.88 total items of job burnoutquestionnaire through the consistent test. All 19 items with three latent factors from burnout questionnaire identified by confirm factor analysis from structure equation model, which RMSEA was 0.088. There was better discriminant validity among occupations.Part II Study into Occupational Stress Health Effect4 Epidemiological study into the relationship between occupationalstress and job bumoutThere were 1186 questionnaires available at this study. Finding showed that there were significant correlation between some factors of JDC model and ERI model and three dimensions of job burnout through hierarchical moderated multiple regression analysis. The factors from ERI model had more power in prediction motional exhaustion and depersonalization and less in reduced personal accomplishment compared to the factors of JDC model. There was 5.1% positive rate of job burnout based on our data to set standard from Maslach burnout theory. It was 5.2% positive rate if Maslach burnout standard. There was dose-action relationship between occupational stress scales from ERI model and emotional exhaustion, depersonalization and burnout by logistic regression analysis. Odd ratios (95% confidence interval) for medium and high group with low as control ranged from 2.76 (1.76～4.07) to 8.71 (5.702～13.31) at emotional exhaustion, from 2.56(1.74～3.75) to 5.78(3.87～8.63) at depersonalization, from 2.78(1.14～6.77) to 3.46(1.40～8.50) at job burnout. High over-commitment people had high risks at emotional exhaustion (OR =2.76, 95%CI 1.72～4.43). High occupational stress group by JDC model had high risk at emotional exhaustion compared to low group (OR=1.63, 95%CI 1.10-2.40). Higher social support had protected from depersonalization and personal accomplishment. Odd ratios (95% confidence interval) at medium and high group equaled to 0.54 (0.38～0.77) and 0.42 (0.29～0.61) for depersonalization, and reached to 0.68 (0.49～0.96) and 0.40 (0.28～0.77) for reduced personal accomplishment. Finding showed combination ERI model with JDC model had more power in predicting job burnout. Factors of ERI model had more power than that of JDC model in predicting job burnout.5 Epidemiological researches into the relationship between occupationalstress and depression trendCenter for Epidemiological Survey Depression Scale (CES-D) from NJH was used to evaluate depression. It was defined as depression trend if whose score answered by CES-D is over twenty. There were 1212 questionnaires qualified for analysis. The positive rate of depression trend was 34.2%. There was a dose-response relation between job stress level by JDC model or ERI model assessment. The score of depression increased within low, medium and high occupational stress groups. Meanwhile, the score of depression reduced at low, medium and high social support group. Finding showed that there was not significant difference between variance occupational types by JDC model by logistic regression analysis. There was significant difference between different job types by ERI model. Odd ratios reached to 2.91(1.93～4.39) at passive job and 3.08(1.98～4.81) at high strain job by low strain job as control. There was not significant difference between active job and low strain job. High over-commitment had more risk in depression trend which odd ratio equaled to 1.77(95%CI: 1.28～2.47). Hobby and better work satisfaction protected from depression trend and odd ratios respectively 0.56(95%CI: 0.41～0.77) and 0.47(95%CI: 0.32～068). There was not significant difference between different job stress ranks by JDC model for depression trend. High social support protected from depression trend (OR=0.67, 95%CI: 0.45～0.98). There was a dose-response relation between depression trend and occupational stress ranks by ERI model assessment. Odd ratio reached to 2.29(95%CI: 1.55～3.39) at medium stress group and 3.18(95%CI: 2.10～4.82) at high stress group. High over-commitment is a risk factor in depression trend (OR=2.39, 95%CI 1.47～3.89). Hobby and better work satisfaction were important protective factors at depression trend by hierarchical moderated multiple regression analysis. ERI model had more power than JDC model in predicting depression trend based on contribution to total variance.6 Investigation on occupational stress and hypertension prevalence riskThere were 695 subjects participating periodic health examination. The information of personal behavior and lifestyle were recollected and occupational stress assessed. Finding showed that hypertension prevalence was 5.9%. There was asignificant relation between hypertension prevalence and age, education level, over-weight (BMI≥24.0), hypertension family history and high occupational stress by JDC model through multiple logistic regression analysis. High occupational stress by JDC model (the rate of job demand and control over 1.0) was an important risk factor in hypertension prevalence. Odd ratio reached to 5.41and 95% confidence interval ranged from 1.18 to 24.82. Over-weight and hypertensive family history were more subjected to high blood pressure (over-weight OR=2.97, 95%CI: 1.39～6.38. hypertension family history OR=4.67, 95%CI: 2.11～10.33). Age over 45 was more risk to hypertension prevalence than age below 30 (OR=6.86, 95%CI: 1.93～24.38). Smoking and drinking were significant difference at single factor analysis but they were none at multiple analysis. High occupational stress may change personal behaviors such as smoking or drinking. The result clue to high occupational stress may be one of important risk factors of hypertension.7 Study into health effect of occupational stress by path analysisTo explore the inner relationship among occupational stress, job burnout, work satisfaction, depression trend and hypertension with path analysis from structure equation model. Six hypothesis models set up based on literatures review and outcome of this data analysis. Hypothesis model was tested by path analysis. Result showed combination model was better one among all hypothesis models according to model fit index. Combination model meant occupation stress comprised by two latent variables (JDC model and ERI model). Job burnout is a latent variable and consisted of depersonalization and reduced personal accomplishment. Two latent variables of occupational stress lead to job burnout, two latent and job burnout combined and acted to satisfaction, depression and hypertension. Absolute model fit index RESMA was 0.129 and goodness fit index was 0.857. This model can be acceptive basically.